I have one bug that I simply cannot track down. Both the attached files are
mostly identical except that m.pgc calls a function before selecting data.
What happens is that both give an identical pointer to ECPGdo but when
create_statement in libecpg reads the data from the stack it only is correct
for n.pgc not for m.pgc.
I tried reproducing this bug with small C sample but failed. That one works
well. But with these two programs it's completely reproducable.
Before I dig into it even more could anyone with a different system and a
different C compiler please try this. Just compile and link with the latest
ecpg from CVS. Can be run against 6.5.3 backend. You just have to make sure
the database exists.
Michael Meskes | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz | Go Rhein Fire!
Tel.: (+49) 2431/72651 | Use Debian GNU/Linux!
Email: Michael(at)Fam-Meskes(dot)De | Use PostgreSQL!
pgsql-hackers by date
|Next:||From: Peter Eisentraut||Date: 2000-04-02 11:33:39|
|Subject: Re: Regression tests|
|Previous:||From: Michael Meskes||Date: 2000-04-02 08:07:50|
|Subject: Re: Call for porting reports|